Video-based Assessment of Preschool Children's Gross Motor Development

NCT07054437 | Recruiting

• 1 other identifier: 2501160032
• Study Type: observational
• Target: 250
• Locations: 1 country
• Sites: 1

Brief Summary

Artificial intelligence (AI) is currently one of the global focal points for industrial development, with its applications in healthcare steadily increasing, such as in disease prediction, image diagnosis, and drug development. AI assists healthcare professionals in clinical decision-making by training relevant models through algorithms, thereby enhancing medical efficiency and quality. Currently, standardized tools are used in clinical settings to screen and assess various aspects of child development. Children's motor development levels are determined by comparing their performance against established norms. However, the current assessment methods primarily rely on on-site visual observation and recording by evaluators, which demands significant time and human resources. This research aims to establish an automated screening tool for gross motor development in early intervention, suitable for independently walking children aged one to six years old in Taiwan. The goal is to reduce the time cost of manual assessment and enable remote healthcare applications.

Conditions

Gross Motor Development Delay

Keywords

artificial intelligence; machine learning; neural networks

Outcome Measures

Primary Outcomes (1)

• Accuracy of AI-based gross motor development screening model compared to pediatric therapist's CDIIT gross motor subscale assessment

  Accuracy will be calculated by comparing the AI model's classification results to pediatric therapists' assessments based on the CDIIT gross motor subscale. The accuracy formula is: (True Positive + True Negative) / Total number of cases.

  Day 1 (single assessment at enrollment).

Interventions

Gross Motor Development Screening Tool OTHER

This intervention is an automated gross motor development screening tool specifically designed for independently walking children aged one to six years old in Taiwan. What sets it apart is its use of artificial intelligence (AI) algorithms to analyze motion data, enabling early identification of potential gross motor developmental delays. Unlike traditional methods that rely on manual, visual observation and subjective recording by healthcare professionals, this tool aims to significantly reduce assessment time and human resource costs. Furthermore, its automated nature makes it uniquely suited for telemedicine applications, allowing for remote screenings and overcoming geographical barriers to access early intervention services. The tool will be developed and validated against established developmental norms relevant to the Taiwanese population.

Eligibility Criteria

• Age: 1 Year - 6 Years
• Sex: all
• Age Groups: Child (0-17)
• Sampling Method: Non-Probability Sample

Study Population

Participants will be recruited from individuals referred for early intervention assessments at hospitals.

You may qualify if:

• Legal guardian willing to provide written informed consent.
• Males and females aged 1 to 6 years old.
• Capable of independent walking.

You may not qualify if:

• \- Non-native Chinese speakers.

Sponsors & Collaborators

• Chang Gung Memorial Hospital: lead
• National Taiwan Normal University: collaborator

Study Sites (1)

Linkou Chang Gung Memorial Hospital

Taoyuan District, Taiwan

RECRUITING

Central Study Contacts

yijing li

CONTACT

+886-920226108; yjl9528@gmail.com

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: CROSS SECTIONAL
• Target Duration: 3 Months
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: June 10, 2025
• First Posted: July 8, 2025
• Study Start: June 11, 2025
• Primary Completion (Estimated): June 11, 2026
• Study Completion (Estimated): August 31, 2026
• Last Updated: July 8, 2025

Record last verified: 2025-06

Data Sharing

• IPD Sharing: Will share

Locations

TW (1)